JIITA, vol.6 no.1, p.513-522, 2022, DOI:
Decision Tree Development with Hierarchical Structure and its Optimization: A solution to overfitting
Yeheng Ma 1), Sanghyuk Lee 1*)
1) Department of Mechatronics and Robotics, School of Advanced Technology, Xi’an Jiaotong-Liverpool University, Suzhou, China
Abstract: A decision tree design is carried out with the hierarchical structure. Specifically, K-fold and random forest algorithm is considered to overcome overfitting problem. Because of the existing problem –overfitting- is the most challenging together with performance when we consider decision tree building. Furthermore, big data and test data decision has always poor prediction together with complex models. One of the reasons could be due to the complex situation in real application cases compared to sample ones for training. The paper propose solutions to the overfitting issues of decision tree models. Simulation results are illustrated.
Keywords: Decision Tree, Overfitting, Random Forest, k-fold validation